Genotyping 80 C57BL/6 GEM from JAX for a mutant nicotinamide nucleotide transhydrogenase (Nnt) allele missing several exons, which is specific to the C57BL/6J substrain, revealed that 27 GEM (34%) actually contained the intact Nnt gene, which is found in all other substrains of C57BL/6 mice. Further studies revealed that the C57BL/6J substrain from JAX was significantly less susceptible to acetaminophen-induced liver injury (AILI) than three other substrains of C57BL/6 mice carrying the intact Nnt gene from other vendors. This finding is very important because it explains the conflicting results in the literature as to the role c-Jun N-terminal kinase 2 (JNK2)-signaling in AILI. In one study, GEM deficient in JNK2 were found to be more susceptible to AILI than control C57BL/6J mice, suggesting that JNK2 had a protective role in AILI. Other researchers concluded that JNK2 was involved in the etiology of AILI because they found that mice deficient in JNK2 were less susceptible to AILI than wild-type C57BL/6 littermates. We discovered that the JNK2 deficient mice from JAX were not on a C57BL/6J background, but instead backcrossed to a C57BL/6 substrain containing the intact Nnt gene that was more susceptible to AILI than the C57BL/6J substrain used as controls in the studies with GEM deficient in JNK2. It is this mispairing that lead to the incorrect conclusion that JNK2 was protective against AILI because the JNK2 deficient mice on a C57BL/6 substrain possessing the intact Nnt gene were inherently more susceptible to AILI than C57BL/6J mice irrespective of the JNK2 gene. Recent genome-wide expression analysis with exon arrays of livers from acetaminophen-treated C57BL/6J mice from JAX and C57BL/6 mice from Taconic containing the intact Nnt gene revealed numerous differences in expressed genes and spliced variants between the two substrains. Conclusion: All biomedical researchers should be aware of the genetic and phenotypic differences in C57BL/6 substrains. This knowledge is particularly important when doing studies with GEM on a C57BL/6 background. Missing pairing of GEM with the wrong C57BL/6 substrain control can lead to erroneous and misleading findings. On the other hand, studying the genetic and phenotypic difference in C57BL/6 substrains may lead to the identification of important risk factors that may play a role in a variety pathological processes.